Everything that uses out2
must be scoped in that function where the console.log
is. That's because JavaScript is asynchronous: when that entire code block runs, all it does is "queue" the HTTP request to be done, and then continues running your code before the request has even been made. All it knows is that whenever it's done with the HTTP request, it will call the two functions passed in in the then
calls, or if it failed, it'll run the function passed into the catch
call.
So the code that does whatever with the result must be a separate function to run maybe later when the request completes. Otherwise, this would just freeze the browser or server until it finishes, and it can't do anything else in the meantime!
If you want a more procedural style, you can use async functions and await in it. For example:
try {
const out = await fetch(...).then(out => out.json())
// ^ note here that those two `out` variables are completely independent
console.log('Fetch Output', out)
} catch (err) {
console.error(err)
}
The same can also be written as this:
try {
const out = await fetch(...)
const out2 = out.json()
console.log('Fetch Output', out)
} catch (err) {
console.error(err)
}
These two examples are just additional syntax sugar to do the same as with the Promise API in your original code. Internally it splits the code into the same sort of nested functions, but it's easier to read and manage especially if you need to wait on many things.
Of course you need to be in an async context for that:
async function doAsyncStuff() {
// Do the above stuff
}
doAsyncStuff().catch(err => console.error(err))
// Here again, the actual execution of `doAsyncStuff` will be put in the queue
// to be ran later, there's no escaping that. But the
// code inside the function can make use of the `await` keyword.